BriskChain: Decentralized Function Composition for High-Performance Serverless Computing

Abstract

Serverless computing allows developers to create workflows for complex tasks through the composition of serverless functions. Current serverless workflow engines rely on master-side pattern which do not permit direct interaction between consecutive serverless functions in the workflow. In this paper, a decentralized worker-side pattern is proposed that provides better performance by allowing consecutive functions to flow from the current node to the next without first having to interact with the master controller. It treats the serverless workflow as a whole unit and uses a locality strategy to optimize performance. This approach is implemented in a workflow engine called BriskChain to demonstrate its effectiveness. Experiments using both synthetic and real application workflows show that BriskChain requires significantly reduced runtime overhead compared with OpenWhisk and Apache Composer, two open-source serverless platforms.